<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2024-revisiting">
    <titleInfo>
        <title>Revisiting Catastrophic Forgetting in Large Language Model Tuning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Hongyu</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Liang</namePart>
        <namePart type="family">Ding</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Meng</namePart>
        <namePart type="family">Fang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dacheng</namePart>
        <namePart type="family">Tao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2024</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yaser</namePart>
            <namePart type="family">Al-Onaizan</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mohit</namePart>
            <namePart type="family">Bansal</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yun-Nung</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Miami, Florida, USA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Catastrophic Forgetting (CF) means models forgetting previously acquired knowledge when learning new data. It compromises the effectiveness of large language models (LLMs) during fine-tuning, yet the underlying causes have not been thoroughly investigated. This paper takes the first step to reveal the direct link between the flatness of the model loss landscape and the extent of CF in the field of LLMs. Based on this, we introduce the sharpness-aware minimization to mitigate CF by flattening the loss landscape. Experiments on three widely-used fine-tuning datasets, spanning different model scales, demonstrate the effectiveness of our method in alleviating CF. Analyses show that we nicely complement the existing anti-forgetting strategies, further enhancing the resistance of LLMs to CF.</abstract>
    <identifier type="citekey">li-etal-2024-revisiting</identifier>
    <identifier type="doi">10.18653/v1/2024.findings-emnlp.249</identifier>
    <location>
        <url>https://aclanthology.org/2024.findings-emnlp.249/</url>
    </location>
    <part>
        <date>2024-11</date>
        <extent unit="page">
            <start>4297</start>
            <end>4308</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Revisiting Catastrophic Forgetting in Large Language Model Tuning
%A Li, Hongyu
%A Ding, Liang
%A Fang, Meng
%A Tao, Dacheng
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Findings of the Association for Computational Linguistics: EMNLP 2024
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F li-etal-2024-revisiting
%X Catastrophic Forgetting (CF) means models forgetting previously acquired knowledge when learning new data. It compromises the effectiveness of large language models (LLMs) during fine-tuning, yet the underlying causes have not been thoroughly investigated. This paper takes the first step to reveal the direct link between the flatness of the model loss landscape and the extent of CF in the field of LLMs. Based on this, we introduce the sharpness-aware minimization to mitigate CF by flattening the loss landscape. Experiments on three widely-used fine-tuning datasets, spanning different model scales, demonstrate the effectiveness of our method in alleviating CF. Analyses show that we nicely complement the existing anti-forgetting strategies, further enhancing the resistance of LLMs to CF.
%R 10.18653/v1/2024.findings-emnlp.249
%U https://aclanthology.org/2024.findings-emnlp.249/
%U https://doi.org/10.18653/v1/2024.findings-emnlp.249
%P 4297-4308

Markdown (Informal)

[Revisiting Catastrophic Forgetting in Large Language Model Tuning](https://aclanthology.org/2024.findings-emnlp.249/) (Li et al., Findings 2024)

• Revisiting Catastrophic Forgetting in Large Language Model Tuning (Li et al., Findings 2024)

ACL

• Hongyu Li, Liang Ding, Meng Fang, and Dacheng Tao. 2024. Revisiting Catastrophic Forgetting in Large Language Model Tuning. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 4297–4308, Miami, Florida, USA. Association for Computational Linguistics.